|
On August 10 2010 07:01 FakeSteve[TPR] wrote:
zerglings aren't a lot weaker than they were in bw. a lot of powerful zerg builds involve ensuring safety with large groups of speedlings. with the advent of creep tumors & improved pathing AI, lings are far stronger relative to bw when it's just zerglings on the field.
Yea well ultimately both the Terran and Zerg armies have improved a lot in SC2 compared to their previous counterparts. And I don't think anyone is trying to say there's a massive 99-1 imbalance of Terran > Zerg, however there is a few key areas in the MU where there is definitely need for further balance tweaks, and quite simply Lings is not the where the problem is. Lings are fine.
|
Idra's 87% win ratio is on the US server. I'm pretty sure he couldn't replicate that on the Korean one (not over a large amount of games, anyway). I recall he had around 67% on his Korean account in the beta.
|
Valhalla18444 Posts
On August 10 2010 07:06 cmos543 wrote:Show nested quote +On August 10 2010 06:58 FakeSteve[TPR] wrote:
intotherainbow didnt play much during beta as far as i know
idra has played more practice games of sc2 than anyone on the planet, his 87%+ win rate on ladder is really very meaningless to a balance discussion. So are you saying that we should just keep comparing IdrA to people in the US server? If Idra's ladder games are meaningless then why does he ladder at all? Why doesnt he just play Koreans in practice games? Why is he even in the US servers when he lives in Korea? I think whats really lacking in this discussion is opening up the field to the data from the Koreans under the assumption that there are at least some that are on par with IdrA that are not zerg, thereby eliminating the variable of skill from the discussion and simplifying the terms.
idra plays ladder games because blizzard has repeatedly stated that top ladder players will be invited to tournaments
that, and the attention! being the top player on the ladder is fun
|
Erm just wanted to mention something: From the 16 or so players in the german ESL Pro Series, 1 Zerg has qualified. haha.
|
On August 10 2010 07:02 SichuanPanda wrote:Show nested quote +On August 10 2010 06:55 cmos543 wrote:On August 10 2010 06:44 SichuanPanda wrote:On August 10 2010 06:40 cmos543 wrote:On August 10 2010 06:17 Apolo wrote:On August 10 2010 06:01 cmos543 wrote:On August 10 2010 05:48 Apolo wrote:On August 10 2010 05:17 Toxigen wrote:
To all those disagreeing with IdrA out there, I'd like you to address his argument for a second:
If Zergs are beating Terrans at high level, it doesn't prove that ZvT is balanced. The other variable is player skill. This means one of two things must be true:
1. Player skill is equal, so if Zerg is winning close to 50%, the races are balanced.
2. Player skill is unequal, so if Zerg is winning close to 50%, the races are imbalanced.
He's also saying that he knows for a fact that Terran players, largely, are often newer to the game compared to Protoss and Zerg players in Korea. His proof stems from his knowledge and experience of those players both in SC1 and SC2. Also, these players were no-names previous to phase 2 (i.e., before mech buffs and roach/infestor nerfs).
These new, no-name, and (in some cases) non-professionals are defeating professional Zergs in tournaments. For #1 above to be true, the following must also be true:
These Terrans are just as skilled as professionals without the practice, knowledge and coaching that those professionals have (i.e., they're simply way more talented than those Zerg & Protoss players). Furthermore, despite playing in phase 1, their skill and talent with the race didn't kick-in until phase 2 and release.
The question becomes: does this sound plausible to you? I agree. I'd like also to add that the statistics can't be taken from the whole population of players in SC2, otherwise results are going to get unreliable. This is because i saw someone analyzing some statistics where the whole population was considered, and that's pretty much useless. Being balanced at bronze level doesn't mean being balanced at the top. Because of that, if the average of winnings is 50-50 throughout all levels, would mean nothing but that the average is 50-50. It could be 60-40 for the bottom and 40-60 for the top skill level and 50-50 inbetween, or the inverse, or more extreme values. I find it very strange that old BW protoss and zerg players are struggling with terran gamers from other RTSes at top tournament levels. That and that i rarely see a terran top player complaining about imbalance of zerg or protoss, while the reverse is so much more common. Anyway, it's blizzard's game to balance. If they don't they're the ones to lose. This whole premise has a simple fallacy that I would like to point out, its so large that I'm very surprised that you don't realize it when its blatantly staring you and IdrA in the face. STATISTICAL SIGNIFICANCE, I mean what heck man. Have you been to highschool? Have IdrA play 50 games against this lessor player and see how the overall win percentages are. You take the 2 games he has played against someone and extrapolate it into way more then it should be without even a single thought. Jesus this is just sad. Well actually i had a chair of statistics at college, which was far more complicated than all highschool classes put together. What you said is a straw man because i didn't mention Idra at all or his 2 games. I based what i said on more than just him. Why did Artosis complain about terran mech, why Idra complained about even more than mech? Why did Whitera and Huk complain about marauders? Why did Tester said Terran was the best race? Why haven't i seen a top terran player complain about zerg or protoss in the last month or two? Why is it that Protoss is by far the most popular race at lower levels, and then the top 100 of diamond Terran takes that place?(1) Why is it that most tournament winnings and runner ups are Terran?(2) I can't comment on the win ratios of the leagues because the B.net system messes up with win ratios directly by trying to make everyone 50-50. And at the top it's even more different. I'll have to think more about that. (1) http://sc2ranks.com/stats/all/1/100(2) there's a post somehwere in this thread that has a big list of tournaments and their respective winners, and most without a doubt are terran. "To all those disagreeing with IdrA out there, I'd like you to address his argument for a second" "I agree." Who knows why the Terrans don't complain, does this automatically mean that there is an imbalance because they are not complaining? I really don't see what concrete evidence you are basing your argument on except for the top 100 which doesn't even support your argument, so far its just been here-say. Even if Terrans are winning the majority of Tournaments, which ones are included? You don't even account for statistical deviation or even attempt. Yes Terrans could be winning 48 out of 50 tournaments but what if no good zergs were in those tournaments(IdrA doesn't feel its worth his time to enter many of the smaller touraments)? Also many tournaments are one round until the semi finals, which makes it harder to get an accurate picture of whats really going on. Maybe someone not so good could get lucky and take a game off of IdrA, knocking him out. But in reality has a very slim chance of beating him in a BO3, especially after he has seen what type of player they are. Both of you have forgotten the real argument altogether and now debating the semantics of statistical analysis. This is unnecessary and totally off topic from what this thread is about. Its really quite simple, at any given level of the T v Z match-up there is imbalances, and there are not nearly all one sided (as in Terran > Zerg imbalance, there is some Zerg > Terran imbalance in some cases). This is a fact that we all need to accept in order to properly discuss and balance the game. You two on the other hand arguing about who's better at statistics, and both refusing to accept there is two way imbalances in the MU are holding back the discussion on a whole. Please stop. Heres something to think about that needs to be said. IdrA is not the only progamer, there are Terran progamers that have even more "advantages" then he does as they can speak Korean and get the full benefit of Korean practice partners. If Terran is so overpowered shouldn't their win ratios reflect that? The Top Korean (I think its into the rainbow but I'm not sure) right now has a win ratio of 66.15 % Idra's is 87.33%. And among Koreans Terran's win ratios look slightly worse from a quick glance. This is the information you should be using, not players that aren't on the same schedule as he is. Stop. Using. IdrA. To. Define. An. Entire. MU. He's a very rare and exceptionally skilled Zerg player that doesn't come around too often. You cannot use his play and his play alone as evidence to Zerg being better. Stop doing it. Everyone here thinks your foolish for doing so, why do you continue? By your logic BW Terran is OP because Flash has an 80 something % win ratio. Clearly that is not the case.
REALLY? Thats exactly WHAT I'm saying infact my posts have been in response to someone doing exactly this, using IdrA to define the MU, doing this seems like all zergs that have lost a games vs mech favorite hobby these days. I guess you didn't read the post I was responding to in the beginning which is understandable.
|
On August 10 2010 07:06 cmos543 wrote:Show nested quote +On August 10 2010 06:58 FakeSteve[TPR] wrote:
intotherainbow didnt play much during beta as far as i know
idra has played more practice games of sc2 than anyone on the planet, his 87%+ win rate on ladder is really very meaningless to a balance discussion. So are you saying that we should just keep comparing IdrA to people in the US server? If Idra's ladder games are meaningless then why does he ladder at all? Why doesnt he just play Koreans in practice games? Why is he even in the US servers when he lives in Korea?
I think whats really lacking in this discussion is opening up the field to the data from the Koreans under the assumption that there are at least some that are on par with IdrA that are not zerg, thereby eliminating the variable of skill from the discussion and simplifying the terms.
You're either retarded or intentionally stupid. Idra's so far above almost everyone on the Us server that using his win rate to gauge balance is stupid is what he's saying. How does this have anything (ANYTHING) to do with why idra ladders at all?
On topic:
I definitely think bunker salvage should either be heavily nerfed or removed entirely, its so stupid. Bunker rush for free, block a zergs expansion for free, defend zerg's all ins for free.
I would prefer nerfed though because it seems like in TvP without making a lot of bunkers, its almost impossible to defend early warpgate pushes from protoss.
Also, blizzard fucking suck at making maps. With the exception of metalopolis and maybe scrap station the maps are a complete joke. Instead of going for balance and gameplay blizzard just crammed every map with a million strategic options in a desperate attempt to scream "lawl guys look how strategic our games is look at every strategic thing you can strategically do in this strategic game its so deep!!111". Seriously....kulas ravine and lost temple....I can't even fathom how ignorant you'd have to be to think a cliff directly beside a mineral line is even remotely balanced.
|
On August 10 2010 07:15 SubtleArt wrote:Show nested quote +On August 10 2010 07:06 cmos543 wrote:On August 10 2010 06:58 FakeSteve[TPR] wrote:
intotherainbow didnt play much during beta as far as i know
idra has played more practice games of sc2 than anyone on the planet, his 87%+ win rate on ladder is really very meaningless to a balance discussion. So are you saying that we should just keep comparing IdrA to people in the US server? If Idra's ladder games are meaningless then why does he ladder at all? Why doesnt he just play Koreans in practice games? Why is he even in the US servers when he lives in Korea?
I think whats really lacking in this discussion is opening up the field to the data from the Koreans under the assumption that there are at least some that are on par with IdrA that are not zerg, thereby eliminating the variable of skill from the discussion and simplifying the terms. You're either retarded or intentionally stupid. Idra's so far above almost everyone on the Us server that using his win rate to gauge balance is stupid is what he's saying. How does this have anything (ANYTHING) to do with why idra ladders at all? On topic: I definitely think bunker salvage should either be heavily nerfed or removed entirely, its so stupid. Bunker rush for free, block a zergs expansion for free, defend zerg's all ins for free. I would prefer nerfed though because it seems like in TvP without making a lot of bunkers, its almost impossible to defend early warpgate pushes from protoss. Also, blizzard fucking suck at making maps. With the exception of metalopolis and maybe scrap station the maps are a complete joke. Instead of going for balance and gameplay blizzard just crammed every map with a million strategic options in a desperate attempt to scream "lawl guys look how strategic our games is look at every strategic thing you can strategically do in this strategic game its so deep!!111". Seriously....kulas ravine and lost temple....I can't even fathom how ignorant you'd have to be to think a cliff directly beside a mineral line is even remotely balanced.
Reread what I wrote and then apologize, I'm circuitously implying what you put so bluntly because I was talking to Fake Steve.
|
|
|
On August 10 2010 05:31 kajeus wrote:Show nested quote +On August 10 2010 05:17 Toxigen wrote:
To all those disagreeing with IdrA out there, I'd like you to address his argument for a second:
If Zergs are beating Terrans at high level, it doesn't prove that ZvT is balanced. The other variable is player skill. This means one of two things must be true:
1. Player skill is equal, so if Zerg is winning close to 50%, the races are balanced.
2. Player skill is unequal, so if Zerg is winning close to 50%, the races are imbalanced.
He's also saying that he knows for a fact that Terran players, largely, are often newer to the game compared to Protoss and Zerg players in Korea. His proof stems from his knowledge and experience of those players both in SC1 and SC2. Also, these players were no-names previous to phase 2 (i.e., before mech buffs and roach/infestor nerfs).
These new, no-name, and (in some cases) non-professionals are defeating professional Zergs in tournaments. For #1 above to be true, the following must also be true:
These Terrans are just as skilled as professionals without the practice, knowledge and coaching that those professionals have (i.e., they're simply way more talented than those Zerg & Protoss players). Furthermore, despite playing in phase 1, their skill and talent with the race didn't kick-in until phase 2 and release.
The question becomes: does this sound plausible to you? I analyzed racial skill levels in this post: http://www.teamliquid.net/forum/viewmessage.php?topic_id=141732¤tpage=15#283 . Feel free to respond in either this thread or that one.
According to your numbers, 20% of players worldwide are Zerg, down 10% from expected (a 30/30/30/10 split for T, P, Z and R, respectively). So worldwide, Terran and Protoss are more popular than Zerg. Popularity counts for a lot -- nobody plays a race they view as weak or that makes them lose more often than they win.
Furthermore, you say that Zerg players have the highest representation in diamond out of their own population percentage-wise. That means that while there might be more Terran players in diamond than Zerg in a strictly numerical sense, diamond players make up more of Zerg's population than Terran's population. It also accounts for the fact that in lower leagues and at lower skill levels, players are more likely to switch races in hopes of better success since they have less practice and experience invested in making Zerg work.
Not only does this data suggest that Zerg is a more difficult race to play (its success is skewed to higher skill levels), but that even then, Zerg is under-represented in top leagues. If I understood it correctly, I'd say that this information corroborates IdrA's hypothesis. I could be wrong.
But really, if zerg is underpowered and they still have such a solid footing in diamond despite their small numbers, that's amazing! Either there are no glaring balance issues or good zerg players are just really fucking good. You summed it up quite nicely with this -- agreed!
EDIT: I think I misread your point above (I missed the "no" initially and thought you were being sarcastic!), so I may have to take back the "agreed"  . I'll leave it in instead of editing it out, though.
I still think that Zerg under-representation in low-skill leagues and over-representation in higher skill leagues is due to the ability to race-switch and the difficulty of Zerg. I mean, just look at the ridiculous over-representation of Terran in North America bronze (46% -- almost half!).
|
We don't have match-up-specific data, but we do have win ratios for each league: http://sc2ranks.com/stats/all/1/all. Check it out. Zergs, terrans, and protoss all do identically well all across diamond. Moreover, the racial distribution of the top 50, 100, and 500 for every region conform precisely to expectations, assuming a random distribution of skill across all players (actually, ONLY if we adjust for the hypothetical higher average skill level of zerg players -- otherwise zergs look way overpowered). So you'll need to be more specific about what exactly you think is not "valid" about these analyses.
There's a danger using win ranks as evidence for imbalance since even if ZvT were imbalanced you wouldn't expect a win ratio much different than 50%, even at the highest skill levels (unless we're discussing gross, obvious imbalance which isn't the case with SC2).
For example, Player A is good, but Player B is slightly better. In a perfect world playing a game like Warcraft II, and based purely on skill/gamesense/APM, Player B beats Player A 70% of the time, maybe a little more.
In SC2, Player A plays Terran and Player B plays Zerg, and Player B only beats A 60% of the time. Knowing nothing about the backstory or identity of these players, and using only this win rank percentage as evidence, you might make the erroneous conclusion that Zerg is overpowered.
EDIT: My implied point was that a poor Terran player might have a 50% win percentage against a better Zerg player, but you'd never know there was a racial imbalance looking at the win percentages.
|
On August 10 2010 07:01 FakeSteve[TPR] wrote:
zerglings aren't a lot weaker than they were in bw. a lot of powerful zerg builds involve ensuring safety with large groups of speedlings. with the advent of creep tumors & improved pathing AI, lings are far stronger relative to bw when it's just zerglings on the field.
Isn't it more accurate to say that lings are more mobile, but more vulnerable since lings no longer have the benefit of SC1's armor system or protection via overkill?
|
On August 10 2010 07:37 Toxigen wrote:Show nested quote +On August 10 2010 05:31 kajeus wrote:On August 10 2010 05:17 Toxigen wrote:
To all those disagreeing with IdrA out there, I'd like you to address his argument for a second:
If Zergs are beating Terrans at high level, it doesn't prove that ZvT is balanced. The other variable is player skill. This means one of two things must be true:
1. Player skill is equal, so if Zerg is winning close to 50%, the races are balanced.
2. Player skill is unequal, so if Zerg is winning close to 50%, the races are imbalanced.
He's also saying that he knows for a fact that Terran players, largely, are often newer to the game compared to Protoss and Zerg players in Korea. His proof stems from his knowledge and experience of those players both in SC1 and SC2. Also, these players were no-names previous to phase 2 (i.e., before mech buffs and roach/infestor nerfs).
These new, no-name, and (in some cases) non-professionals are defeating professional Zergs in tournaments. For #1 above to be true, the following must also be true:
These Terrans are just as skilled as professionals without the practice, knowledge and coaching that those professionals have (i.e., they're simply way more talented than those Zerg & Protoss players). Furthermore, despite playing in phase 1, their skill and talent with the race didn't kick-in until phase 2 and release.
The question becomes: does this sound plausible to you? I analyzed racial skill levels in this post: http://www.teamliquid.net/forum/viewmessage.php?topic_id=141732¤tpage=15#283 . Feel free to respond in either this thread or that one. According to your numbers, 20% of players worldwide are Zerg, down 10% from expected (a 30/30/30/10 split for T, P, Z and R, respectively). So worldwide, Terran and Protoss are more popular than Zerg. Popularity counts for a lot -- nobody plays a race they view as weak or that makes them lose more often than they win.
I'm sorry, but I think you don't know what's meant by "as expected". I'm using statistical language here to mean: "The distribution at the top does not deviate from the distribution in the total population."
Zerg is not very popular. But that does not prove that it's underpowered. Considering that 20% of people worldwide play zerg, it is surprising that 24.43% of diamond league is zerg players. That's a high percentage!
Furthermore, you say that Zerg players have the highest representation in diamond out of their own population percentage-wise. That means that while there might be more Terran players in diamond than Zerg in a strictly numerical sense, diamond players make up more of Zerg's population than Terran's population. It also accounts for the fact that in lower leagues and at lower skill levels, players are more likely to switch races in hopes of better success since they have less practice and experience invested in making Zerg work.
Yes, 6.15% of zergs are in diamond, and 4.5% of terrans are in diamond, as I said. However, zergs are also a higher percentage of the diamond league than they are of the general population. I then proceed to remove the "unexpected" zerg diamond league players (1% of the zerg population) and make them terrans (or whatever). After this adjustment, 19.74% were zerg, which is very similar to the population distribution, in which 20.54% of all players worldwide are zerg. In other words, this is exactly what we would expect.
Not only does this data suggest that Zerg is a more difficult race to play (its success is skewed to higher skill levels), but that even then, Zerg is under-represented in top leagues. If I understood it correctly, I'd say that this information corroborates IdrA's hypothesis. I could be wrong.
How do you conclude that zerg is a more difficult race to play? We have evidence that proportionally more zergs are in diamond league than is true of terran or protoss. It is possible that more bad zergs have moved to protoss or terran, or it's possible that it's just easier for zergs to get into diamond league. Maybe zerg is an easier race to play well enough to get to diamond league! The numbers don't answer that question. All they say is that there are more diamond zergs than expected, and that even if you take those zergs and make them terran, zergs are still as well-represented in diamond league as you would expect them to be if skill were randomly distributed across all players, given their race choices.
|
On August 10 2010 07:46 Toxigen wrote:Show nested quote +We don't have match-up-specific data, but we do have win ratios for each league: http://sc2ranks.com/stats/all/1/all. Check it out. Zergs, terrans, and protoss all do identically well all across diamond. Moreover, the racial distribution of the top 50, 100, and 500 for every region conform precisely to expectations, assuming a random distribution of skill across all players (actually, ONLY if we adjust for the hypothetical higher average skill level of zerg players -- otherwise zergs look way overpowered). So you'll need to be more specific about what exactly you think is not "valid" about these analyses. There's a danger using win ranks as evidence for imbalance since even if ZvT were imbalanced you wouldn't expect a win ratio much different than 50%, even at the highest skill levels (unless we're discussing gross, obvious imbalance which isn't the case with SC2). For example, Player A is good, but Player B is slightly better. In a perfect world playing a game like Warcraft II, and based purely on skill/gamesense/APM, Player B beats Player A 70% of the time, maybe a little more. In SC2, Player A plays Terran and Player B plays Zerg, and Player B only beats A 60% of the time. Knowing nothing about the backstory or identity of these players, and using only this win rank percentage as evidence, you might make the erroneous conclusion that Zerg is overpowered. EDIT: My implied point was that a poor Terran player might have a 50% win percentage against a better Zerg player, but you'd never know there was a racial imbalance looking at the win percentages.
I think the point you're making is that matchmaking will always yield 50/50 win ratios, if it's working. This one seems to work.
That's why we look at the distribution of players who make it to diamond and players in the top 50/100/150/200, etc. If matchmaking yields balanced TvZ results but good zergs are getting paired against worse terrans (because these "worse terrans" do as well, on average, as good zergs), then zergs should be underrepresented in diamond league and the top 50/100/150/200. However, they are not. Zergs are represented as well as you would expect them to be, considering that not a lot of people play them. Zergs are 20% of the total SC2 population, but they are 24.43% of diamond league.
|
I'm sorry, but I think you don't know what's meant by "as expected". I'm using statistical language here to mean: "The distribution at the top does not deviate from the distribution in the total population." Yes, you're using statistical jargon. I'm not. Looking to make sure it's 20% Zerg at the top if there's 20% Zerg overall is much different than asking why it's 20% Zerg in the first place. I think the real question we SHOULD be asking is: why isn't Zerg as popular in SC2 as it was in SC1, (which is where I'm grabbing my 30/30/30/10 split) considering the flavor of the race hasn't changed much (i.e., accounting somewhat for the "coolness" factor). You can't deny that accessibility and imbalance (perceived or otherwise) could skew the population this way. However, I do concede that imbalance and ease of use aren't the only variables here.
I remember seeing a poll on TL.net in the beta asking "Which race will you play in SC2?" It was split pretty evenly between all 3 races. Why isn't the rest of the player pool, minus random players, also split evenly in release? Could imbalance be a factor? That's the thrust of my observation.
How do you conclude that zerg is a more difficult race to play?
There are more Zergs in higher leagues, percentage-wise, than in lower leagues. That means that either:
1. Good players just gravitate to Zerg for some odd reason, while other races have a more normal distribution (not likely).
2. Less-skilled players who initially chose Zerg found them too hard and race-switched to something more manageable.
If a race A is harder to play at a silver level than race B for a given player, isn't that implying imbalance already?
If you make the argument that as the skill of the player increases, Zerg's deficiencies (whatever they are) become more manageable, that might explain the prevalence of diamond Zerg players. However, it doesn't explain why Zerg players are only 20% across the board even in diamond.
I realize that all of this is relatively tangential from my original post, which was imploring TL.net forumers to not discount IdrA's observations simply because they didn't like the personality of the messenger. IdrA's not talking about diamond population percentages -- he's talking about a very select group of (mostly Korean) progamers.
|
Expected value is a phrase with an actual meaning and you're just applying it in cases where it remains to be shown it applies. Why would the percentage of a race out of the total SC2 population be equal to the percentage of that race out of the total population in any given subset of the population stratified by skill level? Why would we expect race popularity to remain constant at all skill levels?
|
On August 10 2010 07:37 Toxigen wrote:Show nested quote +On August 10 2010 05:31 kajeus wrote:On August 10 2010 05:17 Toxigen wrote:
To all those disagreeing with IdrA out there, I'd like you to address his argument for a second:
If Zergs are beating Terrans at high level, it doesn't prove that ZvT is balanced. The other variable is player skill. This means one of two things must be true:
1. Player skill is equal, so if Zerg is winning close to 50%, the races are balanced.
2. Player skill is unequal, so if Zerg is winning close to 50%, the races are imbalanced.
He's also saying that he knows for a fact that Terran players, largely, are often newer to the game compared to Protoss and Zerg players in Korea. His proof stems from his knowledge and experience of those players both in SC1 and SC2. Also, these players were no-names previous to phase 2 (i.e., before mech buffs and roach/infestor nerfs).
These new, no-name, and (in some cases) non-professionals are defeating professional Zergs in tournaments. For #1 above to be true, the following must also be true:
These Terrans are just as skilled as professionals without the practice, knowledge and coaching that those professionals have (i.e., they're simply way more talented than those Zerg & Protoss players). Furthermore, despite playing in phase 1, their skill and talent with the race didn't kick-in until phase 2 and release.
The question becomes: does this sound plausible to you? I analyzed racial skill levels in this post: http://www.teamliquid.net/forum/viewmessage.php?topic_id=141732¤tpage=15#283 . Feel free to respond in either this thread or that one. According to your numbers, 20% of players worldwide are Zerg, down 10% from expected (a 30/30/30/10 split for T, P, Z and R, respectively). So worldwide, Terran and Protoss are more popular than Zerg. Popularity counts for a lot -- nobody plays a race they view as weak or that makes them lose more often than they win. Furthermore, you say that Zerg players have the highest representation in diamond out of their own population percentage-wise. That means that while there might be more Terran players in diamond than Zerg in a strictly numerical sense, diamond players make up more of Zerg's population than Terran's population. It also accounts for the fact that in lower leagues and at lower skill levels, players are more likely to switch races in hopes of better success since they have less practice and experience invested in making Zerg work. Not only does this data suggest that Zerg is a more difficult race to play (its success is skewed to higher skill levels), but that even then, Zerg is under-represented in top leagues. If I understood it correctly, I'd say that this information corroborates IdrA's hypothesis. I could be wrong. Show nested quote +But really, if zerg is underpowered and they still have such a solid footing in diamond despite their small numbers, that's amazing! Either there are no glaring balance issues or good zerg players are just really fucking good. You summed it up quite nicely with this -- agreed!
Popularity is a decent gauge of balance, but it's far from absolute, many other factors come into play when someone decides what race they want to play, and while perceived strength might be one of them, a ton of other factors will come into play. The game very well could be 30/30/30/10 at any given level and could still be unbalanced, coming into these discussions and pointing to just the representation stats and then asking blizzard to fix this isn't helping them, at all.
Not saying there might not be some imbalances, but many players are dealing in absolutes or have broken down and have stopped trying to innovate in favor of playing the back seat devs.
|
You're way overanalyzing. The reason a higher % of Zerg is in diamond league is because Zerg vs the other matchups is balanced or even in favour of Zerg with players who have no macro and micro skills. But once Terrans properly harass without losing their harassing units due to mis-microing or once they start using effective drops and get their timings straight, TvZ is heavily in favour of T.
But why is that so? Well that's relatively simple: While the possiblities with other races open up as the player of the race gets better, Zerg options are always very limited as we simply have the fewest amount of units and are a very reactive race overall. Shortly said, Diamond Zergs and zergs in lower leagues largely do the same things as we have no other options. But on lower leagues what we do is not effectively countered, while at diamond level it is. The problem with the whole matchup is that we have no "re-counters" in early game against terran.
|
United Kingdom12027 Posts
There are more Zergs in higher leagues, percentage-wise, than in lower leagues. That means that either:
1. Good players just gravitate to Zerg for some odd reason, while other races have a more normal distribution (not likely).
2. Less-skilled players who initially chose Zerg found them too hard and race-switched to something more manageable.
Using that logic, you could also say that Zerg are very easy to play and get to high levels with. 
|
On August 10 2010 08:44 heishe wrote:
You're way overanalyzing. The reason a higher % of Zerg is in diamond league is because Zerg vs the other matchups is balanced or even in favour of Zerg with players who have no macro and micro skills. But once Terrans properly harass without losing their harassing units due to mis-microing or once they start using effective drops and get their timings straight, TvZ is heavily in favour of T.
But why is that so? Well that's relatively simple: While the possiblities with other races open up as the player of the race gets better, Zerg options are always very limited as we simply have the fewest amount of units and are a very reactive race overall. Shortly said, Diamond Zergs and zergs in lower leagues largely do the same things as we have no other options. But on lower leagues what we do is not effectively countered, while at diamond level it is. The problem with the whole matchup is that we have no "re-counters" in early game against terran.
I think you're overanalyzing. If there's a higher percent of a race, that means that race is ahead. But yes, there is a slight lean to Terran, which is largely due to Terran having more options in the form of more units.
|
On August 10 2010 08:40 tetracycloide wrote:
Expected value is a phrase with an actual meaning and you're just applying it in cases where it remains to be shown it applies. Why would the percentage of a race out of the total SC2 population be equal to the percentage of that race out of the total population in any given subset of the population stratified by skill level? Why would we expect race popularity to remain constant at all skill levels?
Because we're assuming a random distribution of skill across races. The leagues are split into skill level, and we're assuming a constant (or near-constant) percentage of each race will make it to diamond league. If 5% of players overall make it into diamond league, we would expect 5% of zergs, 5% of terrans, and 5% of protoss to make it to diamond league.
However, 6.15% of zergs make it to diamond league. That's why I adjusted for the "unexpected 1.15%" in my post on the topic, to which we've been linking throughout. Even after making this 1.15% into terran players (i.e., adjusting for the hypothetically superior average skill of zergs), the distribution of diamond players is still consistent with the distribution of all players worldwide.
In other words, even assuming that zergs are better than terrans on average, zergs still show up in diamond exactly as much as you'd expect them to -- in exactly the same proportion as is present in the worldwide population.
|
|
|
|
|
|
EPT![[image loading]](http://img267.imageshack.us/i/idrarage.png/)
